Dehydration of Amides to Nitriles under Conditions of a Catalytic Appel Reaction

Article abstract of DOI:10.1021/acs.orglett.7b03862

A highly expedient protocol for a catalytic Appel-type dehydration of amides to nitriles has been developed that employs oxalyl chloride and triethylamine along with triphenylphosphine oxide as a catalyst. The reactions are usually complete in less than 10 min with only a 1 mol % catalyst loading. The reaction scope includes aromatic, heteroaromatic, and aliphatic amides, including derivatives of α-hydroxy and α-amino acids.

Full text of DOI:10.1021/acs.orglett.7b03862

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
Dehydration of Amides to Nitriles under Conditions of a Catalytic
Appel Reaction
Sergei A. Shipilovskikh,† Vladimir Yu. Vaganov,† Elena I. Denisova,† Aleksandr E. Rubtsov,*,†
and Andrei V. Malkov*,‡,§
†Department of Chemistry, Perm State University, Bukireva 15, Perm 614990, Russia
‡Department of Chemistry, Loughborough University, Loughborough, Leicestershire LE11 3TU, U.K.
§Department of Organic Chemistry, RUDN University, 6 Miklukho-Maklaya Street, Moscow 117198, Russia
S
* Supporting Information
ABSTRACT: A highly expedient protocol for a catalytic
Appel-type dehydration of amides to nitriles has been
developed that employs oxalyl chloride and triethylamine
along with triphenylphosphine oxide as a catalyst. The
reactions are usually complete in less than 10 min with only
a 1 mol % catalyst loading. The reaction scope includes
aromatic, heteroaromatic, and aliphatic amides, including
derivatives of α-hydroxy and α-amino acids.
hemical compounds with the nitrile functional group
serve as useful precursors in the manufacture of a large
Scheme 1. Dehydration of Amides to Nitriles by Appel
Reaction
C
variety of consumer products, such as polyamides, pigments
and dyes, pharmaceuticals, agrochemicals, and many other
substances.1 The nitrile group has a rich chemistry since it can
be conveniently converted to other functional groups.2 Also,
the nitrile pharmacophore plays a significant role in modulating
the biological activity of synthetic medicinal drugs3 and natural
products.4
The synthesis of nitriles by dehydration of primary amides is
a well-known process; numerous methods have been developed
and used by practicing chemists for over a century.5
In the past few years, a number of new methods have
emerged for carrying out efficient dehydration of amides to
nitriles. Among the new reagents, [Et2NSF2]BF4 (XtalFluor-E)6
and silicotungstate [n-Bu4N]4[α-H4SiW11O39]7 have been
introduced. Also, metal-catalyzed dehydration has been carried
out in the presence of silanes,8 N-methyl-N-(trimethylsilyl)-
trifluoroacetamide,9 and acetonitrile.10
However, in many instances the practicality of the methods is
offset by the toxicity of the reagents, long reaction times,
narrow scope, and the laborious workup needed to remove the
side products. Therefore, the development of convenient,
practical methods for converting amides to nitriles still holds its
relevance.
quantities of Ph3PO, which complicates the isolation and
purification of the target products. Therefore, the development
of a catalytic version of the Appel dehydration of amides in
which CCl4 as a source of chloride is replaced by a safer
alternative and the phosphorus reagent is used as an organic
catalyst while the universal application scope of the original
method is kept would make it significantly more attractive from
the practical standpoint. Our work was inspired by the recent
reports of Denton14 and others,15 who have demonstrated that
running some reactions under the Appel conditions can be
efficient using catalytic Ph3PO in the presence of oxalyl
The Appel reaction,11 which employs a combination of Ph3P,
CCl4, and Et3N, belongs to a group of highly versatile tools
that, among other useful transformations,12 can be used for the
synthesis of nitriles from amides (Scheme 1A). However, the
original Appel protocol has a number of weaknesses. First of all,
exposure to CCl4 can lead to kidney damage, and it is also a
suspected carcinogen. Furthermore, it contributes to depletion
of the ozone layer and can no longer be used.13 Second,
stoichiometric use of Ph3P results in the formation of large
Received: December 11, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.7b03862
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
chloride. In this work, we present an adaptation of this protocol
to the mild and facile conversion of amides to nitriles using as
low as a 1 mol % loading of Ph3PO (Scheme 1B).
The investigation commenced with establishing the best
conditions for the dehydration of amides, employing
benzamide (1a) as a model substrate (Table 1). First, the
conversion (entry 4), which increased to 29% and 45% when
triethylamine was used in 2- and 3-fold excess, respectively
(entries 5 and 6). Full conversion was achieved when the
amount of (COCl)2 was doubled to 2 equiv (entry 7). It should
be noted that the reaction was complete in less than 10 min. A
3-fold excess of triethylamine appears to be crucial, as reducing
the amount of Et3N to 2 equiv slowed the reaction (entry 8).
Next, the effect of the catalyst loading was investigated
(Table 1, entries 9−11). Pleasingly, Ph3PO performed
efficiently even when the loading was as low as 1 mol %
(entry 11). With 1a the catalyst content could be further
reduced, but for investigating the reaction scope, 1 mol %
proved to be most suitable. The influence of the solvent was
briefly assessed (entries 12−15). Despite some good
conversions achieved in Et2O and CH2Cl2 (entries 12 and
13), MeCN remained the best choice. Other solvents, such as
toluene and dioxane, proved to be inferior (entries 14 and 15).
Thus, the conditions in entry 11 were taken as optimal.
a
Table 1. Optimization of the Reaction Conditions
equiv of
(COCl)2
mol %
Ph3PO
equiv of
Et3N
conv.
b
entry
solvent
(%)
c
1
2
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
−
−
−
1
−
1
2
3
3
2
3
3
3
3
3
3
3
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
Et2O
0
d
0
3
100
10
10
10
10
10
5
25
14
4
To investigate the reaction scope, the substrates were
selected to represent diverse compound classes. The results
are collected in Figure 1. On a standard 2 mmol scale, 2a was
5
29
6
45
7
100
75
8
9
100
100
100
87
10
11
12
13
14
15
2
1
1
1
CH2Cl2
toluene
dioxane
85
1
77
1
31
a
General conditions: 1a (0.25 mmol), Ph3PO, Et3N, dry MeCN (1
mL), dropwise addition of neat (COCl)2. The reactions were carried
out for 10 min at rt before an aliquot (50 μL) was taken, quenched
b
with aqueous MeCN (1 mL), and analyzed by GC. Conversion to 2a
c
was calculated from GC. After 10 min, according to GC, the reaction
mixture contained 1a (70%) and benzoyl isocyanate (3a) (ca. 25%).
d
After 10 min, according to GC, the reaction mixture contained 1a
(60%), 3a (ca. 21%), and some unidentified byproducts.
role of each reagent was evaluated. Oxalyl chloride on its own
did not produce benzonitrile (2a) (entry 1). Instead, slow
formation of benzoyl isocyanate (3a) was observed, mirroring
published data.16 This reaction is likely to proceed through
initial O-acylation of 1a to give oxalyl imidate 4 followed by
cyclization to form unstable 2-phenyloxazolin-4,5-dione (5),
which then forms 3a (Scheme 2). Addition of triethylamine to
Scheme 2. Formation of Benzoyl Isocyanate (3a)
the reaction mixture did not alter the reaction outcome (entry
2), which contrasts with the previously reported formation of
2a when 1a was treated with oxalyl chloride in the presence of
2,6-lutidine.16a
The use of stoichiometric quantities of Ph3PO and (COCl)2
without triethylamine resulted in low conversion of 1a into 2a
(Table 1, entry 3). With 10 mol % Ph3PO and 1 equiv each of
oxalyl chloride and trimethylamine, 2a was formed in 14%
Figure 1. Catalytic dehydration scope. General conditions: amide 1
(2.0 mmol), Ph3PO (1 mol %), Et3N (3 equiv), dry MeCN (8 mL),
dropwise addition of (COCl)2 (2 equiv). The reactions were carried
out for 10 min at rt. Yields of the isolated compounds are given.
B
DOI: 10.1021/acs.orglett.7b03862
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
isolated in 98% yield. The reaction worked equally well on a 1 g
(8.3 mmol) scale, giving 2a in 92% yield. Aromatic amides 1b−
f mirrored the reactivity of the model amide 1a, furnishing the
respective nitriles 2b−f in high yields. A set of heterocyclic
amides 1g−j were next assessed. 2-Cyanothiophene (2g), 3-
cyanopyridine (2h), and 2-cyanoquinoline (2i) were accessed
in high yields. In this series, it is worth highlighting the efficient
synthesis of 3-cyanotetrahydroisoquinoline derivative 2j from
the corresponding amide 1j. This compound serves as a
precursor to the corresponding axially chiral tetrahydroisoqui-
noline-N,N′-dioxide, which belongs to a group of highly
competent Lewis base catalysts for asymmetric crotylation of
carbonyl compounds that are currently under investigation in
our laboratories.17 Unsaturated cinnamamide (1k) and aliphatic
linear (1l, 1m) and branched (1n, 1o) amides were converted
into the respective nitriles in good yields, though their reactivity
was somewhat lower compared with the aromatic analogues.
The same applied to the functionalized derivatives 1p−r, which
included phenoxyacetamide (1p) and phthalimide-protected
amino acid amides 1q and 1r. It has to be noted that the
corresponding Boc-protected amino acids did not give the
desired nitrile. Instead, numerous side products were formed,
which most likely resulted from the reaction of the N-Boc
group with oxalyl chloride, similar to the reaction shown in
Scheme 2.
MeCN. After 10 min at rt, conversion to nitrile 2a reached
30%. However, when Et3N was used in a combination with 1
equiv of (COCl)2, complete conversion to 2a was achieved
almost instantaneously, thus making this route highly probable.
Currently, by monitoring the reaction by NMR spectroscopy,
we were unable to distinguish whether it proceeds through the
intermediacy of D or F (or both) because of the short reaction
times. Therefore, both catalytic cycles could be considered
viable options.19
In conclusion, we have developed a highly expedient protocol
for catalytic Appel-type dehydration of amides to nitriles
catalyzed by triphenylphosphine oxide. The salient features of
the method are (i) operationally simplicity, (ii) low catalyst
loading (1 mol %), (iii) short reaction times, (iv) mild
conditions, and (v) wide reaction scope that includes aromatic,
heteroaromatic, and aliphatic amides, including functionalized
derivatives.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.7b03862.
1
Experimental procedures and H and 13C NMR spectra
for new compounds (PDF)
On the basis of the experimental observations and the
published data on the catalytic Appel reaction,14,15 we suggest
two possible mechanistic pathways for this dehydration reaction
(Scheme 3). Both cycles start with quick formation of the
AUTHOR INFORMATION
Corresponding Authors
■
*E-mail: A.Malkov@lboro.ac.uk.
Scheme 3. Possible Catalytic Cycles for Dehydration of
Amides to Nitriles Mediated by Triphenylphosphine Oxide
*E-mail: Rubtsov@psu.ru.
ORCID
Sergei A. Shipilovskikh: 0000-0002-8917-2583
Aleksandr E. Rubtsov: 0000-0002-4299-3464
Andrei V. Malkov: 0000-0001-6072-2353
Author Contributions
The manuscript was written through contributions of all
authors.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors thank the Russian Science Foundation for Grant
15-13-00092. We also acknowledge RUDN University and the
Ministry of Education and Science of the Russian Federation
(Agreement 02.A03.21.0008) for access to analytical facilities
(HRMS).
intermediate chlorophosphonium chloride (B) upon treatment
of triphenylphosphine oxide (A) with (COCl)2, as demon-
strated by Denton with the aid of 31P NMR spectroscopy.14d
Next, in catalytic cycle 1, intermediate B, in the presence of
Et3N, reacts with amide 1 via oxygen to form first species D
followed by species E, which then undergoes elimination to
furnish nitrile 2 and regenerate A; such a sequence was
proposed by Appel.11 The pathway outlined in catalytic cycle 2,
which proceeds via formation of N-acyltriphenylphosphine
imide F, was originally rejected by Appel11 because F was
deemed to be too stable. Nonetheless, we decided to revisit this
mechanistic option (see the Supporting Information for
details). N-Benzoyltriphenylphosphine imide synthesized by a
literature method18 was first treated with 3 equiv of Et3N in
REFERENCES
■
(1) (a) Fatiadi, A. J. Preparation and synthetic applications of cyano
compounds. In Triple-Bonded Functional Groups; Patai, S., Rappoport,
Z., Eds.; John Wiley & Sons: Chichester, U.K., 1983; Vol. 2. (b) Miller,
J. S.; Manson, J. L. Acc. Chem. Res. 2001, 34, 563−570.
(2) (a) Gaspar, B.; Carreira, E. M. Angew. Chem. 2007, 119, 4603−
4606. (b) Kukushkin, V. Y.; Pombeiro, A. J. L. Chem. Rev. 2002, 102,
1771−1802.
(3) Fleming, F. F.; Yao, L.; Ravikumar, P. C.; Funk, L.; Shook, B. C. J.
Med. Chem. 2010, 53, 7902−7917.
(4) Fleming, F. F. Nat. Prod. Rep. 1999, 16, 597−606.
(5) (a) Smith, M. B.; March, J. Eliminations. In March’s Advanced
Organic Chemistry: Reactions, Mechanisms, and Structure, 6th ed.; John
Wiley & Sons: Hoboken, NJ, 2006; pp 1549−1550. (b) Bhattacharyya,
C
DOI: 10.1021/acs.orglett.7b03862
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
N. K.; Jha, S.; Jha, S.; Bhutia, T. Y.; Adhikary, G. Int. J. Chem. Appl.
2012, 4, 295−304.
(6) Keita, M.; Vandamme, M.; Paquin, J.-F. Synthesis 2015, 47,
3758−3766.
(7) Itagaki, S.; Kamata, K.; Yamaguchi, K.; Mizuno, N. ChemCatCh-
em 2013, 5, 1725−1728.
(8) (a) Elangovan, S.; Quintero-Duque, S.; Dorcet, V.; Roisnel, T.;
Norel, L.; Darcel, C.; Sortais, J.-B. Organometallics 2015, 34, 4521−
4528. (b) Zhou, S.; Addis, D.; Das, S.; Junge, K.; Beller, M. Chem.
Commun. 2009, 45, 4883−4885. (c) Xue, B.; Sun, H.; Wang, Y.;
Zheng, T.; Li, X.; Fuhr, O.; Fenske, D. Catal. Commun. 2016, 86, 148−
150.
(9) (a) Enthaler, S.; Inoue, S. Chem. - Asian J. 2012, 7, 169−175.
(b) Enthaler, S. Chem. - Eur. J. 2011, 17, 9316−9319.
(10) Zhang, W.; Haskins, C. W.; Yang, Y.; Dai, M. Org. Biomol. Chem.
2014, 12, 9109−9112.
(11) Appel, R.; Kleinstuck, R.; Ziehn, K. D. Chem. Ber. 1971, 104,
̈
1030−1034.
(12) (a) Appel, R. Angew. Chem., Int. Ed. Engl. 1975, 14, 801−811.
(b) Lee, R. J.; Lindley, M. R.; Pritchard, G. J.; Kimber, M. C. Chem.
Commun. 2017, 53, 6327−6330.
(13) Dow, K.; Downing, T. E. An Atmospheric Life Time of 85 Years
Has Been Ascribed to CCl4. In The Atlas of Climate Change; University
of California Press: Berkeley, CA, 2006.
(14) (a) Denton, R. M.; An, J.; Adeniran, B. Chem. Commun. 2010,
46, 3025−3027. (b) Denton, R. M.; Tang, X.; Przeslak, A. Org. Lett.
2010, 12, 4678−4681. (c) An, J.; Denton, R. M.; Lambert, T. H.;
Nacsa, E. D. Org. Biomol. Chem. 2014, 12, 2993−3003. (d) Denton, R.
M.; An, J.; Adeniran, B.; Blake, A. J.; Lewis, W.; Poulton, A. M. J. Org.
Chem. 2011, 76, 6749−6767.
(15) (a) Rajendran, K. V.; Gilheany, D. G. Chem. Commun. 2012, 48,
817−819. (b) Nikitin, K.; Muller-Bunz, H.; Gilheany, D. Chem.
̈
Commun. 2013, 49, 1434−1436. (c) Yu, T. Y.; Wang, Y.; Xu, P. F.
Chem. - Eur. J. 2014, 20, 98−101. (d) Kenny, N. P.; Rajendran, K. V.;
Gilheany, D. G. Chem. Commun. 2015, 51, 16561−16564.
(16) (a) Couturier, M.; Caron, L.; Tumidajski, S.; Jones, K.; White,
T. D. Org. Lett. 2006, 8, 1929−1932. (b) Speziale, A. J.; Smith, L. R. J.
Org. Chem. 1963, 28, 1805−1811.
(17) (a) Incerti-Pradillos, C. A.; Kabeshov, M. A.; O’Hora, P. S.;
Shipilovskikh, S. A.; Rubtsov, A. E.; Drobkova, V. A.; Balandina, S. Y.;
Malkov, A. V. Chem. - Eur. J. 2016, 22, 14390−14396. (b) O’Hora, P.
S.; Incerti-Pradillos, C. A.; Kabeshov, M. A.; Shipilovskikh, S. A.;
Rubtsov, A. E.; Elsegood, M. R.; Malkov, A. V. Chem. - Eur. J. 2015, 21,
4551−4555.
(18) Frøyen, P. Phosphorus, Sulfur Silicon Relat. Elem. 1993, 78, 161−
171.
(19) One of the reviewers suggested another possible mechanistic
scenario for the dehydration of amides using the same set of reagents:
imidate 4 formed by O-acylation of 1 (Scheme 2) reacts with
triphenylphosphine oxide to give the active intermediate D, which is
then deprotonated by triethylamine to yield 2 (Scheme 3). While we
cannot rule out such a possibility, our optimization experiments (Table
1, entries 1 and 2) suggest that if 4 were formed in appreciable
quantities, we should have observed formation of the respective acyl
isocyanates, which were not detected under the optimized reaction
conditions. Therefore, this route appears to be less likely than the two
shown in Scheme 3.
D
DOI: 10.1021/acs.orglett.7b03862
Org. Lett. XXXX, XXX, XXX−XXX